Patchwork [1/2] RTC: Add core rtc support for Gemini Soc devices

login
register
mail settings
Submitter Hans Ulli Kroll
Date Dec. 14, 2010, 3:08 p.m.
Message ID <1292339307-14336-2-git-send-email-ulli.kroll@googlemail.com>
Download mbox | patch
Permalink /patch/75493/
State Superseded
Headers show

Comments

Hans Ulli Kroll - Dec. 14, 2010, 3:08 p.m.
driver for the rtc device
on Cortina Systems CS3516 or StormlinkSemi SL3516 aka Gemini SoC

Signed-off-by: Hans Ulli Kroll <ulli.kroll@googlemail.com>
---
 MAINTAINERS              |    1 +
 drivers/rtc/Kconfig      |    9 ++
 drivers/rtc/Makefile     |    1 +
 drivers/rtc/rtc-gemini.c |  206 ++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 217 insertions(+), 0 deletions(-)
 create mode 100644 drivers/rtc/rtc-gemini.c
Alexandre Belloni - May 1, 2015, 11:42 p.m.
Hi,

On 14/12/2010 at 16:08:26 +0100, Hans Ulli Kroll wrote :
> driver for the rtc device
> on Cortina Systems CS3516 or StormlinkSemi SL3516 aka Gemini SoC
> 
> Signed-off-by: Hans Ulli Kroll <ulli.kroll@googlemail.com>

This driver has never been merged and the platform doesn't seem to be
active anymore. Is there still any interest in getting this driver
mainlined?

Only tree wide cleanups happened in mach-gemini since end of 2010, the
listed git repository (git://git.berlios.de/gemini-board) was on berliOS
(closed since 2011) and the sourceforge mirror seems empty. Is there
still interest in keeping that platform in the mainline?

> ---
>  MAINTAINERS              |    1 +
>  drivers/rtc/Kconfig      |    9 ++
>  drivers/rtc/Makefile     |    1 +
>  drivers/rtc/rtc-gemini.c |  206 ++++++++++++++++++++++++++++++++++++++++++++++
>  4 files changed, 217 insertions(+), 0 deletions(-)
>  create mode 100644 drivers/rtc/rtc-gemini.c
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 0094224..1987d46 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -619,6 +619,7 @@ L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
>  T:	git git://git.berlios.de/gemini-board
>  S:	Maintained
>  F:	arch/arm/mach-gemini/
> +F:	drivers/rtc/rtc-gemini.c
>  
>  ARM/EBSA110 MACHINE SUPPORT
>  M:	Russell King <linux@arm.linux.org.uk>
> diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
> index 2883428..da14253 100644
> --- a/drivers/rtc/Kconfig
> +++ b/drivers/rtc/Kconfig
> @@ -839,6 +839,15 @@ config RTC_DRV_BFIN
>  	  This driver can also be built as a module. If so, the module
>  	  will be called rtc-bfin.
>  
> +config RTC_DRV_GEMINI
> +	tristate "Gemini SoC RTC"
> +	help
> +	  If you say Y here you will get support for the
> +	  RTC found on Gemini SoC's.
> +
> +	  This driver can also be built as a module. If so, the module
> +	  will be called rtc-gemini.
> +
>  config RTC_DRV_RS5C313
>  	tristate "Ricoh RS5C313"
>  	depends on SH_LANDISK
> diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile
> index 4c2832d..779e42c 100644
> --- a/drivers/rtc/Makefile
> +++ b/drivers/rtc/Makefile
> @@ -46,6 +46,7 @@ obj-$(CONFIG_RTC_DRV_DS3234)	+= rtc-ds3234.o
>  obj-$(CONFIG_RTC_DRV_EFI)	+= rtc-efi.o
>  obj-$(CONFIG_RTC_DRV_EP93XX)	+= rtc-ep93xx.o
>  obj-$(CONFIG_RTC_DRV_FM3130)	+= rtc-fm3130.o
> +obj-$(CONFIG_RTC_DRV_GEMINI)	+= rtc-gemini.o
>  obj-$(CONFIG_RTC_DRV_GENERIC)	+= rtc-generic.o
>  obj-$(CONFIG_RTC_DRV_IMXDI)	+= rtc-imxdi.o
>  obj-$(CONFIG_RTC_DRV_ISL1208)	+= rtc-isl1208.o
> diff --git a/drivers/rtc/rtc-gemini.c b/drivers/rtc/rtc-gemini.c
> new file mode 100644
> index 0000000..111bb67
> --- /dev/null
> +++ b/drivers/rtc/rtc-gemini.c
> @@ -0,0 +1,206 @@
> +/*
> + *  Gemini OnChip RTC
> + *
> + *  Copyright (C) 2009 Janos Laube <janos.dev@gmail.com>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, write to the Free Software Foundation, Inc.,
> + * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
> + *
> + * Original code for older kernel 2.6.15 are form Stormlinksemi
> + * first update from Janos Laube for > 2.6.29 kernels
> + *
> + * checkpatch fixes and usage off rtc-lib code
> + * Hans Ulli Kroll <ulli.kroll@googlemail.com>
> + */
> +
> +#include <linux/rtc.h>
> +#include <linux/io.h>
> +#include <linux/slab.h>
> +#include <linux/platform_device.h>
> +
> +#include <mach/hardware.h>
> +
> +#define DRV_NAME        "rtc-gemini"
> +#define DRV_VERSION     "0.2"
> +
> +struct gemini_rtc {
> +	struct rtc_device	*rtc_dev;
> +	void __iomem		*rtc_base;
> +	int			rtc_irq;
> +};
> +
> +enum gemini_rtc_offsets {
> +	GEMINI_RTC_SECOND	= 0x00,
> +	GEMINI_RTC_MINUTE	= 0x04,
> +	GEMINI_RTC_HOUR		= 0x08,
> +	GEMINI_RTC_DAYS		= 0x0C,
> +	GEMINI_RTC_ALARM_SECOND	= 0x10,
> +	GEMINI_RTC_ALARM_MINUTE	= 0x14,
> +	GEMINI_RTC_ALARM_HOUR	= 0x18,
> +	GEMINI_RTC_RECORD	= 0x1C,
> +	GEMINI_RTC_CR		= 0x20
> +};
> +
> +static irqreturn_t gemini_rtc_interrupt(int irq, void *dev)
> +{
> +	return IRQ_HANDLED;
> +}
> +
> +/*
> + * Looks like the RTC in the Gemini SoC is (totaly) broken
> + * We can't read/write directly the time from RTC registers.
> + * We must do some "offset" calculation to get the real time
> + *
> + * This FIX works pretty fine and Stormlinksemi aka Cortina-Networks does
> + * the same thing, without the rtc-lib.c calls.
> + */
> +
> +static int gemini_rtc_read_time(struct device *dev, struct rtc_time *tm)
> +{
> +	struct gemini_rtc *rtc = dev_get_drvdata(dev);
> +
> +	unsigned int  days, hour, min, sec;
> +	unsigned long offset, time;
> +
> +	sec  = ioread32(rtc->rtc_base + GEMINI_RTC_SECOND);
> +	min  = ioread32(rtc->rtc_base + GEMINI_RTC_MINUTE);
> +	hour = ioread32(rtc->rtc_base + GEMINI_RTC_HOUR);
> +	days = ioread32(rtc->rtc_base + GEMINI_RTC_DAYS);
> +	offset = ioread32(rtc->rtc_base + GEMINI_RTC_RECORD);
> +
> +	time = offset + days * 86400 + hour * 3600 + min * 60 + sec;
> +
> +	rtc_time_to_tm(time, tm);
> +
> +	return 0;
> +}
> +
> +static int gemini_rtc_set_time(struct device *dev, struct rtc_time *tm)
> +{
> +	struct gemini_rtc *rtc = dev_get_drvdata(dev);
> +	unsigned int sec, min, hour, day;
> +	unsigned long offset, time;
> +
> +	if (tm->tm_year >= 2148)	/* EPOCH Year + 179 */
> +		return -EINVAL;
> +
> +	rtc_tm_to_time(tm , &time);
> +
> +	sec = ioread32(rtc->rtc_base + GEMINI_RTC_SECOND);
> +	min = ioread32(rtc->rtc_base + GEMINI_RTC_MINUTE);
> +	hour = ioread32(rtc->rtc_base + GEMINI_RTC_HOUR);
> +	day = ioread32(rtc->rtc_base + GEMINI_RTC_DAYS);
> +
> +	offset = time - (day*86400 + hour*3600 + min*60 + sec);
> +
> +	iowrite32(offset, rtc->rtc_base + GEMINI_RTC_RECORD);
> +	iowrite32(0x01, rtc->rtc_base + GEMINI_RTC_CR);
> +
> +	return 0;
> +}
> +
> +static struct rtc_class_ops gemini_rtc_ops = {
> +	.read_time     = gemini_rtc_read_time,
> +	.set_time      = gemini_rtc_set_time,
> +};
> +
> +static int __devinit gemini_rtc_probe(struct platform_device *pdev)
> +{
> +	struct gemini_rtc *rtc;
> +	struct device *dev = &pdev->dev;
> +	struct resource *res;
> +	int ret;
> +
> +	rtc = kzalloc(sizeof(*rtc), GFP_KERNEL);
> +	if (unlikely(!rtc))
> +		return -ENOMEM;
> +	platform_set_drvdata(pdev, rtc);
> +
> +	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
> +	if (!res) {
> +		ret = -ENODEV;
> +		goto err_mem;
> +	}
> +	rtc->rtc_irq = res->start;
> +
> +	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> +	if (!res) {
> +		ret = -ENODEV;
> +		goto err_mem;
> +	}
> +	rtc->rtc_base =	devm_ioremap(&pdev->dev, res->start,
> +				res->end - res->start + 1);
> +
> +	ret = request_irq(rtc->rtc_irq, gemini_rtc_interrupt,
> +				IRQF_SHARED, pdev->name, dev);
> +	if (unlikely(ret))
> +		goto err_mem;
> +
> +	rtc->rtc_dev =	rtc_device_register(pdev->name, dev,
> +				&gemini_rtc_ops, THIS_MODULE);
> +	if (unlikely(IS_ERR(rtc->rtc_dev))) {
> +		ret = PTR_ERR(rtc->rtc_dev);
> +		goto err_irq;
> +	}
> +	return 0;
> +
> +err_irq:
> +	free_irq(rtc->rtc_irq, dev);
> +
> +err_mem:
> +	kfree(rtc);
> +	return ret;
> +}
> +
> +static int __devexit gemini_rtc_remove(struct platform_device *pdev)
> +{
> +	struct gemini_rtc *rtc = platform_get_drvdata(pdev);
> +	struct device *dev = &pdev->dev;
> +
> +	free_irq(rtc->rtc_irq, dev);
> +	rtc_device_unregister(rtc->rtc_dev);
> +	platform_set_drvdata(pdev, NULL);
> +	kfree(rtc);
> +
> +	return 0;
> +}
> +
> +static struct platform_driver gemini_rtc_driver = {
> +	.driver		= {
> +		.name	= "rtc-gemini",
> +		.owner	= THIS_MODULE,
> +	},
> +	.probe		= gemini_rtc_probe,
> +	.remove		= __devexit_p(gemini_rtc_remove),
> +};
> +
> +static int __init gemini_rtc_init(void)
> +{
> +	return platform_driver_register(&gemini_rtc_driver);
> +}
> +
> +static void __exit gemini_rtc_exit(void)
> +{
> +	platform_driver_unregister(&gemini_rtc_driver);
> +}
> +
> +module_init(gemini_rtc_init);
> +module_exit(gemini_rtc_exit);
> +
> +MODULE_AUTHOR("Hans Ulli Kroll <ulli.kroll@googlemail.com>");
> +MODULE_DESCRIPTION("RTC driver for Gemini SoC");
> +MODULE_LICENSE("GPL");
> +MODULE_VERSION(DRV_VERSION);
> +MODULE_ALIAS("platform:" DRV_NAME);
> +
Arnd Bergmann - May 6, 2015, 9:39 a.m.
On Saturday 02 May 2015 01:42:14 Alexandre Belloni wrote:
> Hi,
> 
> On 14/12/2010 at 16:08:26 +0100, Hans Ulli Kroll wrote :
> > driver for the rtc device
> > on Cortina Systems CS3516 or StormlinkSemi SL3516 aka Gemini SoC
> > 
> > Signed-off-by: Hans Ulli Kroll <ulli.kroll@googlemail.com>
> 
> This driver has never been merged and the platform doesn't seem to be
> active anymore. Is there still any interest in getting this driver
> mainlined?
> 
> Only tree wide cleanups happened in mach-gemini since end of 2010, the
> listed git repository (git://git.berlios.de/gemini-board) was on berliOS
> (closed since 2011) and the sourceforge mirror seems empty. Is there
> still interest in keeping that platform in the mainline?

As far as I know, the platform is still used by a number of people,
and is supported by OpenWRT. The reason we haven't seen updates is that
Ulli has been mostly absent from upstream development, and we haven't
had any other person step up as maintainer.

I have a patch to convert the platform to ARCH_MULTIPLATFORM, and
the code doesn't really get in the way otherwise.

As far as I'm concerned, we should just merge all the patches from
http://git.openwrt.org/?p=openwrt.git;a=tree;f=target/linux/gemini/patches-3.18

We should also try to find a maintainer that can respond to patches
in a timely manner. If Ulli has time for that again, that would be great,
otherwise I think we should find someone from OpenWRT to take over.

	Arnd
Alexandre Belloni - May 6, 2015, 9:53 a.m.
On 06/05/2015 at 11:39:58 +0200, Arnd Bergmann wrote :
> On Saturday 02 May 2015 01:42:14 Alexandre Belloni wrote:
> > Hi,
> > 
> > On 14/12/2010 at 16:08:26 +0100, Hans Ulli Kroll wrote :
> > > driver for the rtc device
> > > on Cortina Systems CS3516 or StormlinkSemi SL3516 aka Gemini SoC
> > > 
> > > Signed-off-by: Hans Ulli Kroll <ulli.kroll@googlemail.com>
> > 
> > This driver has never been merged and the platform doesn't seem to be
> > active anymore. Is there still any interest in getting this driver
> > mainlined?
> > 
> > Only tree wide cleanups happened in mach-gemini since end of 2010, the
> > listed git repository (git://git.berlios.de/gemini-board) was on berliOS
> > (closed since 2011) and the sourceforge mirror seems empty. Is there
> > still interest in keeping that platform in the mainline?
> 
> As far as I know, the platform is still used by a number of people,
> and is supported by OpenWRT. The reason we haven't seen updates is that
> Ulli has been mostly absent from upstream development, and we haven't
> had any other person step up as maintainer.
> 
> I have a patch to convert the platform to ARCH_MULTIPLATFORM, and
> the code doesn't really get in the way otherwise.
> 
> As far as I'm concerned, we should just merge all the patches from
> http://git.openwrt.org/?p=openwrt.git;a=tree;f=target/linux/gemini/patches-3.18
> 
> We should also try to find a maintainer that can respond to patches
> in a timely manner. If Ulli has time for that again, that would be great,
> otherwise I think we should find someone from OpenWRT to take over.
> 

Thanks to point that, I'll merge that one which seems clean enough:
http://git.openwrt.org/?p=openwrt.git;a=blob;f=target/linux/gemini/files/drivers/rtc/rtc-gemini.c;h=587d8122b2fbb1230437eadcce4789a53aa60ee5;hb=4c637410a2a1ab45988e8ca6202554a502102039
Roman Yeryomin - May 6, 2015, 4:21 p.m.
On 2015-05-06 12:53, Alexandre Belloni wrote:
> On 06/05/2015 at 11:39:58 +0200, Arnd Bergmann wrote :
>> On Saturday 02 May 2015 01:42:14 Alexandre Belloni wrote:
>> > Hi,
>> >
>> > On 14/12/2010 at 16:08:26 +0100, Hans Ulli Kroll wrote :
>> > > driver for the rtc device
>> > > on Cortina Systems CS3516 or StormlinkSemi SL3516 aka Gemini SoC
>> > >
>> > > Signed-off-by: Hans Ulli Kroll <ulli.kroll@googlemail.com>
>> >
>> > This driver has never been merged and the platform doesn't seem to be
>> > active anymore. Is there still any interest in getting this driver
>> > mainlined?
>> >
>> > Only tree wide cleanups happened in mach-gemini since end of 2010, the
>> > listed git repository (git://git.berlios.de/gemini-board) was on berliOS
>> > (closed since 2011) and the sourceforge mirror seems empty. Is there
>> > still interest in keeping that platform in the mainline?
>> 
>> As far as I know, the platform is still used by a number of people,
>> and is supported by OpenWRT. The reason we haven't seen updates is 
>> that
>> Ulli has been mostly absent from upstream development, and we haven't
>> had any other person step up as maintainer.
>> 
>> I have a patch to convert the platform to ARCH_MULTIPLATFORM, and
>> the code doesn't really get in the way otherwise.
>> 
>> As far as I'm concerned, we should just merge all the patches from
>> http://git.openwrt.org/?p=openwrt.git;a=tree;f=target/linux/gemini/patches-3.18
>> 
>> We should also try to find a maintainer that can respond to patches
>> in a timely manner. If Ulli has time for that again, that would be 
>> great,
>> otherwise I think we should find someone from OpenWRT to take over.
>> 
> 
> Thanks to point that, I'll merge that one which seems clean enough:
> http://git.openwrt.org/?p=openwrt.git;a=blob;f=target/linux/gemini/files/drivers/rtc/rtc-gemini.c;h=587d8122b2fbb1230437eadcce4789a53aa60ee5;hb=4c637410a2a1ab45988e8ca6202554a502102039

For 3.18 (and up) to work on gemini 160-gemini-timers.patch is vital. 
Without that timers are broken and cpu is super slow.

Regards,
Roman
Arnd Bergmann - May 6, 2015, 7:22 p.m.
On Wednesday 06 May 2015 19:21:21 Roman Yeryomin wrote:
> 
> > Thanks to point that, I'll merge that one which seems clean enough:
> > http://git.openwrt.org/?p=openwrt.git;a=blob;f=target/linux/gemini/files/drivers/rtc/rtc-gemini.c;h=587d8122b2fbb1230437eadcce4789a53aa60ee5;hb=4c637410a2a1ab45988e8ca6202554a502102039
> 
> For 3.18 (and up) to work on gemini 160-gemini-timers.patch is vital. 
> Without that timers are broken and cpu is super slow.

Do you know what part of the patch is the actual bug fix? We should probably
merge that separately and mark it for stable backports, while the bulk of that
patch seems to just rearrange code.

	Arnd
Hans Ulli Kroll - May 7, 2015, 3 a.m.
Hello

On Wed, 6 May 2015, Arnd Bergmann wrote:

> On Saturday 02 May 2015 01:42:14 Alexandre Belloni wrote:
> > Hi,
> > 
> > On 14/12/2010 at 16:08:26 +0100, Hans Ulli Kroll wrote :
> > > driver for the rtc device
> > > on Cortina Systems CS3516 or StormlinkSemi SL3516 aka Gemini SoC
> > > 
> > > Signed-off-by: Hans Ulli Kroll <ulli.kroll@googlemail.com>
> > 
> > This driver has never been merged and the platform doesn't seem to be
> > active anymore. Is there still any interest in getting this driver
> > mainlined?
> > 
> > Only tree wide cleanups happened in mach-gemini since end of 2010, the
> > listed git repository (git://git.berlios.de/gemini-board) was on berliOS
> > (closed since 2011) and the sourceforge mirror seems empty. Is there
> > still interest in keeping that platform in the mainline?
> 
> As far as I know, the platform is still used by a number of people,
> and is supported by OpenWRT. The reason we haven't seen updates is that
> Ulli has been mostly absent from upstream development, and we haven't
> had any other person step up as maintainer.
> 
> I have a patch to convert the platform to ARCH_MULTIPLATFORM, and
> the code doesn't really get in the way otherwise.
> 
> As far as I'm concerned, we should just merge all the patches from
> http://git.openwrt.org/?p=openwrt.git;a=tree;f=target/linux/gemini/patches-3.18
y> 
> We should also try to find a maintainer that can respond to patches
> in a timely manner. If Ulli has time for that again, that would be great,
> otherwise I think we should find someone from OpenWRT to take over.
> 
> 	Arnd
> 

I'm back in busyness.
Currently me dev device is right next to me.
I moved some time ago and haav to find it in some 'hidden' box.


Ulli
Arnd Bergmann - May 7, 2015, 7:49 a.m.
On Thursday 07 May 2015 05:00:03 Hans Ulli Kroll wrote:
> On Wed, 6 May 2015, Arnd Bergmann wrote:
> > As far as I'm concerned, we should just merge all the patches from
> > http://git.openwrt.org/?p=openwrt.git;a=tree;f=target/linux/gemini/patches-3.18
> > 
> > We should also try to find a maintainer that can respond to patches
> > in a timely manner. If Ulli has time for that again, that would be great,
> > otherwise I think we should find someone from OpenWRT to take over.
> 
> I'm back in busyness.
> Currently me dev device is right next to me.
> I moved some time ago and haav to find it in some 'hidden' box.

Awesome, welcome back!

Are you able to look at the patches I mentioned above and submit them
to arm@kernel.org, cc linux-arm-kernel@lists.infradead.org with proper
patch descriptions?

	Arnd
Roman Yeryomin - May 7, 2015, 9:03 a.m.
On 2015-05-06 22:22, Arnd Bergmann wrote:
> On Wednesday 06 May 2015 19:21:21 Roman Yeryomin wrote:
>> 
>> > Thanks to point that, I'll merge that one which seems clean enough:
>> > http://git.openwrt.org/?p=openwrt.git;a=blob;f=target/linux/gemini/files/drivers/rtc/rtc-gemini.c;h=587d8122b2fbb1230437eadcce4789a53aa60ee5;hb=4c637410a2a1ab45988e8ca6202554a502102039
>> 
>> For 3.18 (and up) to work on gemini 160-gemini-timers.patch is vital.
>> Without that timers are broken and cpu is super slow.
> 
> Do you know what part of the patch is the actual bug fix? We should 
> probably
> merge that separately and mark it for stable backports, while the bulk 
> of that
> patch seems to just rearrange code.
> 
> 	Arnd

Sorry, didn't try to extract the exact lines it but I guess it's that 
part which touches the scheduler clock code.
Because without that patch sched clock runs at 100Hz instead of 25MHz.
I can dive into this deeper if you want, I just didn't see the need.

Regards,
Roman
Hans Ulli Kroll - May 7, 2015, 5:35 p.m.
Hi

On Thu, 7 May 2015, Roman Yeryomin wrote:

> On 2015-05-06 22:22, Arnd Bergmann wrote:
> > On Wednesday 06 May 2015 19:21:21 Roman Yeryomin wrote:
> > > 
> > > > Thanks to point that, I'll merge that one which seems clean enough:
> > > > http://git.openwrt.org/?p=openwrt.git;a=blob;f=target/linux/gemini/files/drivers/rtc/rtc-gemini.c;h=587d8122b2fbb1230437eadcce4789a53aa60ee5;hb=4c637410a2a1ab45988e8ca6202554a502102039
> > > 
> > > For 3.18 (and up) to work on gemini 160-gemini-timers.patch is vital.
> > > Without that timers are broken and cpu is super slow.
> > 
> > Do you know what part of the patch is the actual bug fix? We should probably
> > merge that separately and mark it for stable backports, while the bulk of
> > that
> > patch seems to just rearrange code.
> > 
> >  Arnd
> 
> Sorry, didn't try to extract the exact lines it but I guess it's that part
> which touches the scheduler clock code.
> Because without that patch sched clock runs at 100Hz instead of 25MHz.
> I can dive into this deeper if you want, I just didn't see the need.
> 
> Regards,
> Roman
> 
> 

I've send the RTC driver for Gemini SoC upstream-

Ulli
Hans Ulli Kroll - May 7, 2015, 5:46 p.m.
Hi

On Thu, 7 May 2015, Arnd Bergmann wrote:

> On Thursday 07 May 2015 05:00:03 Hans Ulli Kroll wrote:
> > On Wed, 6 May 2015, Arnd Bergmann wrote:
> > > As far as I'm concerned, we should just merge all the patches from
> > > http://git.openwrt.org/?p=openwrt.git;a=tree;f=target/linux/gemini/patches-3.18
> > > 
> > > We should also try to find a maintainer that can respond to patches
> > > in a timely manner. If Ulli has time for that again, that would be great,
> > > otherwise I think we should find someone from OpenWRT to take over.
> > 
> > I'm back in busyness.
> > Currently me dev device is right next to me.
> > I moved some time ago and haav to find it in some 'hidden' box.
> 
> Awesome, welcome back!
> 
> Are you able to look at the patches I mentioned above and submit them
> to arm@kernel.org, cc linux-arm-kernel@lists.infradead.org with proper
> patch descriptions?
> 
> 	Arnd
> 

Already done,
RTC driver send.
Compilder fix for MACH-RECORD send.
Maintainer Update send.

This device is a (very) bad design, to be polite !
I think they didn't test this properly.
And they fixed some software issues in hardware.
To be exact :
They switched the IDE channels in hardware, to get around the stupid 
Sausalito software.
Roman Yeryomin - May 7, 2015, 9:31 p.m.
On 2015-05-07 20:35, Hans Ulli Kroll wrote:
> Hi
> 
> On Thu, 7 May 2015, Roman Yeryomin wrote:
> 
>> On 2015-05-06 22:22, Arnd Bergmann wrote:
>> > On Wednesday 06 May 2015 19:21:21 Roman Yeryomin wrote:
>> > >
>> > > > Thanks to point that, I'll merge that one which seems clean enough:
>> > > > http://git.openwrt.org/?p=openwrt.git;a=blob;f=target/linux/gemini/files/drivers/rtc/rtc-gemini.c;h=587d8122b2fbb1230437eadcce4789a53aa60ee5;hb=4c637410a2a1ab45988e8ca6202554a502102039
>> > >
>> > > For 3.18 (and up) to work on gemini 160-gemini-timers.patch is vital.
>> > > Without that timers are broken and cpu is super slow.
>> >
>> > Do you know what part of the patch is the actual bug fix? We should probably
>> > merge that separately and mark it for stable backports, while the bulk of
>> > that
>> > patch seems to just rearrange code.
>> >
>> >  Arnd
>> 
>> Sorry, didn't try to extract the exact lines it but I guess it's that 
>> part
>> which touches the scheduler clock code.
>> Because without that patch sched clock runs at 100Hz instead of 25MHz.
>> I can dive into this deeper if you want, I just didn't see the need.
>> 
>> Regards,
>> Roman
>> 
>> 
> 
> I've send the RTC driver for Gemini SoC upstream-

I was talking about 160-gemini-timers.patch, not rtc driver.

Regards,
Roman
Roman Yeryomin - May 7, 2015, 9:35 p.m.
On 2015-05-07 20:46, Hans Ulli Kroll wrote:
> Hi
> 
> On Thu, 7 May 2015, Arnd Bergmann wrote:
> 
>> On Thursday 07 May 2015 05:00:03 Hans Ulli Kroll wrote:
>> > On Wed, 6 May 2015, Arnd Bergmann wrote:
>> > > As far as I'm concerned, we should just merge all the patches from
>> > > http://git.openwrt.org/?p=openwrt.git;a=tree;f=target/linux/gemini/patches-3.18
>> > >
>> > > We should also try to find a maintainer that can respond to patches
>> > > in a timely manner. If Ulli has time for that again, that would be great,
>> > > otherwise I think we should find someone from OpenWRT to take over.
>> >
>> > I'm back in busyness.
>> > Currently me dev device is right next to me.
>> > I moved some time ago and haav to find it in some 'hidden' box.
>> 
>> Awesome, welcome back!
>> 
>> Are you able to look at the patches I mentioned above and submit them
>> to arm@kernel.org, cc linux-arm-kernel@lists.infradead.org with proper
>> patch descriptions?
>> 
>> 	Arnd
>> 
> 
> Already done,
> RTC driver send.
> Compilder fix for MACH-RECORD send.
> Maintainer Update send.
> 
> This device is a (very) bad design, to be polite !
> I think they didn't test this properly.
> And they fixed some software issues in hardware.
> To be exact :
> They switched the IDE channels in hardware, to get around the stupid
> Sausalito software.

AFAIK, ide channels are switchable in the driver.

Regards,
Roman
Arnd Bergmann - May 8, 2015, 4:08 p.m.
On Thursday 07 May 2015 12:03:25 Roman Yeryomin wrote:
> On 2015-05-06 22:22, Arnd Bergmann wrote:
> > On Wednesday 06 May 2015 19:21:21 Roman Yeryomin wrote:
> >> 
> >> > Thanks to point that, I'll merge that one which seems clean enough:
> >> > http://git.openwrt.org/?p=openwrt.git;a=blob;f=target/linux/gemini/files/drivers/rtc/rtc-gemini.c;h=587d8122b2fbb1230437eadcce4789a53aa60ee5;hb=4c637410a2a1ab45988e8ca6202554a502102039
> >> 
> >> For 3.18 (and up) to work on gemini 160-gemini-timers.patch is vital.
> >> Without that timers are broken and cpu is super slow.
> > 
> > Do you know what part of the patch is the actual bug fix? We should 
> > probably
> > merge that separately and mark it for stable backports, while the bulk 
> > of that
> > patch seems to just rearrange code.
> > 
> >       Arnd
> 
> Sorry, didn't try to extract the exact lines it but I guess it's that 
> part which touches the scheduler clock code.
> Because without that patch sched clock runs at 100Hz instead of 25MHz.
> I can dive into this deeper if you want, I just didn't see the need.

I think that would be helpful, yes.

	Arnd
Hans Ulli Kroll - May 8, 2015, 6:28 p.m.
Hi

On Fri, 8 May 2015, Roman Yeryomin wrote:

> On 2015-05-07 20:46, Hans Ulli Kroll wrote:
> > Hi
> > 
> > On Thu, 7 May 2015, Arnd Bergmann wrote:
> > 
> > > On Thursday 07 May 2015 05:00:03 Hans Ulli Kroll wrote:
> > > > On Wed, 6 May 2015, Arnd Bergmann wrote:
> > > > > As far as I'm concerned, we should just merge all the patches from
> > > > > http://git.openwrt.org/?p=openwrt.git;a=tree;f=target/linux/gemini/patches-3.18
> > > > >
> > > > > We should also try to find a maintainer that can respond to patches
> > > > > in a timely manner. If Ulli has time for that again, that would be
> > > > > great,
> > > > > otherwise I think we should find someone from OpenWRT to take over.
> > > >
> > > > I'm back in busyness.
> > > > Currently me dev device is right next to me.
> > > > I moved some time ago and haav to find it in some 'hidden' box.
> > > 
> > > Awesome, welcome back!
> > > 
> > > Are you able to look at the patches I mentioned above and submit them
> > > to arm@kernel.org, cc linux-arm-kernel@lists.infradead.org with proper
> > > patch descriptions?
> > > 
> > >  Arnd
> > > 
> > 
> > Already done,
> > RTC driver send.
> > Compilder fix for MACH-RECORD send.
> > Maintainer Update send.
> > 
> > This device is a (very) bad design, to be polite !
> > I think they didn't test this properly.
> > And they fixed some software issues in hardware.
> > To be exact :
> > They switched the IDE channels in hardware, to get around the stupid
> > Sausalito software.
> 
> AFAIK, ide channels are switchable in the driver.

Yes.
But there is a HW switch.
/* Values of IOMUX
 * 26:24 bits is "IDE IO Select"
 * 111:100 - Reserved
 * 011 - ata0 <-> sata0, sata1; bring out ata1
 * 010 - ata1 <-> sata1, sata0; bring out ata0
 * 001 - ata0 <-> sata0, ata1 <-> sata1; bring out ata1
 * 000 - ata0 <-> sata0, ata1 <-> sata1; bring out ata0
*/

This is from an email long time ago, with Paulius Zaleckas who has the 
yoriginal documentation to this SoC.
I hacked some weeks (or months) to get this thing running.

Some HDD are working well, others don't. At this time, this was only one 
manufacturer, who has only the "bad" devices.
Early this year, I got some report from one user. He has two hdd's from 
the same manufacturer. One hdd is working well, the other don't.

At the end of this "session" I've taken this device to some hidden place.
;-)

Ulli

Patch

diff --git a/MAINTAINERS b/MAINTAINERS
index 0094224..1987d46 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -619,6 +619,7 @@  L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
 T:	git git://git.berlios.de/gemini-board
 S:	Maintained
 F:	arch/arm/mach-gemini/
+F:	drivers/rtc/rtc-gemini.c
 
 ARM/EBSA110 MACHINE SUPPORT
 M:	Russell King <linux@arm.linux.org.uk>
diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
index 2883428..da14253 100644
--- a/drivers/rtc/Kconfig
+++ b/drivers/rtc/Kconfig
@@ -839,6 +839,15 @@  config RTC_DRV_BFIN
 	  This driver can also be built as a module. If so, the module
 	  will be called rtc-bfin.
 
+config RTC_DRV_GEMINI
+	tristate "Gemini SoC RTC"
+	help
+	  If you say Y here you will get support for the
+	  RTC found on Gemini SoC's.
+
+	  This driver can also be built as a module. If so, the module
+	  will be called rtc-gemini.
+
 config RTC_DRV_RS5C313
 	tristate "Ricoh RS5C313"
 	depends on SH_LANDISK
diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile
index 4c2832d..779e42c 100644
--- a/drivers/rtc/Makefile
+++ b/drivers/rtc/Makefile
@@ -46,6 +46,7 @@  obj-$(CONFIG_RTC_DRV_DS3234)	+= rtc-ds3234.o
 obj-$(CONFIG_RTC_DRV_EFI)	+= rtc-efi.o
 obj-$(CONFIG_RTC_DRV_EP93XX)	+= rtc-ep93xx.o
 obj-$(CONFIG_RTC_DRV_FM3130)	+= rtc-fm3130.o
+obj-$(CONFIG_RTC_DRV_GEMINI)	+= rtc-gemini.o
 obj-$(CONFIG_RTC_DRV_GENERIC)	+= rtc-generic.o
 obj-$(CONFIG_RTC_DRV_IMXDI)	+= rtc-imxdi.o
 obj-$(CONFIG_RTC_DRV_ISL1208)	+= rtc-isl1208.o
diff --git a/drivers/rtc/rtc-gemini.c b/drivers/rtc/rtc-gemini.c
new file mode 100644
index 0000000..111bb67
--- /dev/null
+++ b/drivers/rtc/rtc-gemini.c
@@ -0,0 +1,206 @@ 
+/*
+ *  Gemini OnChip RTC
+ *
+ *  Copyright (C) 2009 Janos Laube <janos.dev@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Original code for older kernel 2.6.15 are form Stormlinksemi
+ * first update from Janos Laube for > 2.6.29 kernels
+ *
+ * checkpatch fixes and usage off rtc-lib code
+ * Hans Ulli Kroll <ulli.kroll@googlemail.com>
+ */
+
+#include <linux/rtc.h>
+#include <linux/io.h>
+#include <linux/slab.h>
+#include <linux/platform_device.h>
+
+#include <mach/hardware.h>
+
+#define DRV_NAME        "rtc-gemini"
+#define DRV_VERSION     "0.2"
+
+struct gemini_rtc {
+	struct rtc_device	*rtc_dev;
+	void __iomem		*rtc_base;
+	int			rtc_irq;
+};
+
+enum gemini_rtc_offsets {
+	GEMINI_RTC_SECOND	= 0x00,
+	GEMINI_RTC_MINUTE	= 0x04,
+	GEMINI_RTC_HOUR		= 0x08,
+	GEMINI_RTC_DAYS		= 0x0C,
+	GEMINI_RTC_ALARM_SECOND	= 0x10,
+	GEMINI_RTC_ALARM_MINUTE	= 0x14,
+	GEMINI_RTC_ALARM_HOUR	= 0x18,
+	GEMINI_RTC_RECORD	= 0x1C,
+	GEMINI_RTC_CR		= 0x20
+};
+
+static irqreturn_t gemini_rtc_interrupt(int irq, void *dev)
+{
+	return IRQ_HANDLED;
+}
+
+/*
+ * Looks like the RTC in the Gemini SoC is (totaly) broken
+ * We can't read/write directly the time from RTC registers.
+ * We must do some "offset" calculation to get the real time
+ *
+ * This FIX works pretty fine and Stormlinksemi aka Cortina-Networks does
+ * the same thing, without the rtc-lib.c calls.
+ */
+
+static int gemini_rtc_read_time(struct device *dev, struct rtc_time *tm)
+{
+	struct gemini_rtc *rtc = dev_get_drvdata(dev);
+
+	unsigned int  days, hour, min, sec;
+	unsigned long offset, time;
+
+	sec  = ioread32(rtc->rtc_base + GEMINI_RTC_SECOND);
+	min  = ioread32(rtc->rtc_base + GEMINI_RTC_MINUTE);
+	hour = ioread32(rtc->rtc_base + GEMINI_RTC_HOUR);
+	days = ioread32(rtc->rtc_base + GEMINI_RTC_DAYS);
+	offset = ioread32(rtc->rtc_base + GEMINI_RTC_RECORD);
+
+	time = offset + days * 86400 + hour * 3600 + min * 60 + sec;
+
+	rtc_time_to_tm(time, tm);
+
+	return 0;
+}
+
+static int gemini_rtc_set_time(struct device *dev, struct rtc_time *tm)
+{
+	struct gemini_rtc *rtc = dev_get_drvdata(dev);
+	unsigned int sec, min, hour, day;
+	unsigned long offset, time;
+
+	if (tm->tm_year >= 2148)	/* EPOCH Year + 179 */
+		return -EINVAL;
+
+	rtc_tm_to_time(tm , &time);
+
+	sec = ioread32(rtc->rtc_base + GEMINI_RTC_SECOND);
+	min = ioread32(rtc->rtc_base + GEMINI_RTC_MINUTE);
+	hour = ioread32(rtc->rtc_base + GEMINI_RTC_HOUR);
+	day = ioread32(rtc->rtc_base + GEMINI_RTC_DAYS);
+
+	offset = time - (day*86400 + hour*3600 + min*60 + sec);
+
+	iowrite32(offset, rtc->rtc_base + GEMINI_RTC_RECORD);
+	iowrite32(0x01, rtc->rtc_base + GEMINI_RTC_CR);
+
+	return 0;
+}
+
+static struct rtc_class_ops gemini_rtc_ops = {
+	.read_time     = gemini_rtc_read_time,
+	.set_time      = gemini_rtc_set_time,
+};
+
+static int __devinit gemini_rtc_probe(struct platform_device *pdev)
+{
+	struct gemini_rtc *rtc;
+	struct device *dev = &pdev->dev;
+	struct resource *res;
+	int ret;
+
+	rtc = kzalloc(sizeof(*rtc), GFP_KERNEL);
+	if (unlikely(!rtc))
+		return -ENOMEM;
+	platform_set_drvdata(pdev, rtc);
+
+	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (!res) {
+		ret = -ENODEV;
+		goto err_mem;
+	}
+	rtc->rtc_irq = res->start;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		ret = -ENODEV;
+		goto err_mem;
+	}
+	rtc->rtc_base =	devm_ioremap(&pdev->dev, res->start,
+				res->end - res->start + 1);
+
+	ret = request_irq(rtc->rtc_irq, gemini_rtc_interrupt,
+				IRQF_SHARED, pdev->name, dev);
+	if (unlikely(ret))
+		goto err_mem;
+
+	rtc->rtc_dev =	rtc_device_register(pdev->name, dev,
+				&gemini_rtc_ops, THIS_MODULE);
+	if (unlikely(IS_ERR(rtc->rtc_dev))) {
+		ret = PTR_ERR(rtc->rtc_dev);
+		goto err_irq;
+	}
+	return 0;
+
+err_irq:
+	free_irq(rtc->rtc_irq, dev);
+
+err_mem:
+	kfree(rtc);
+	return ret;
+}
+
+static int __devexit gemini_rtc_remove(struct platform_device *pdev)
+{
+	struct gemini_rtc *rtc = platform_get_drvdata(pdev);
+	struct device *dev = &pdev->dev;
+
+	free_irq(rtc->rtc_irq, dev);
+	rtc_device_unregister(rtc->rtc_dev);
+	platform_set_drvdata(pdev, NULL);
+	kfree(rtc);
+
+	return 0;
+}
+
+static struct platform_driver gemini_rtc_driver = {
+	.driver		= {
+		.name	= "rtc-gemini",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= gemini_rtc_probe,
+	.remove		= __devexit_p(gemini_rtc_remove),
+};
+
+static int __init gemini_rtc_init(void)
+{
+	return platform_driver_register(&gemini_rtc_driver);
+}
+
+static void __exit gemini_rtc_exit(void)
+{
+	platform_driver_unregister(&gemini_rtc_driver);
+}
+
+module_init(gemini_rtc_init);
+module_exit(gemini_rtc_exit);
+
+MODULE_AUTHOR("Hans Ulli Kroll <ulli.kroll@googlemail.com>");
+MODULE_DESCRIPTION("RTC driver for Gemini SoC");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
+MODULE_ALIAS("platform:" DRV_NAME);
+